NeuroCalm by BK

NeuroCalm is an adaptive EdTech tool that detects user stress and delivers real-time, personalized micro-meditation recommendations to help learners build lasting calm. Inspired by the challenges of maintaining focus and well-being in fast-paced digital learning environments, it uses a combination of signal processing and reinforcement learning to identify stress and recommend effective techniques. On first use, the system calibrates a user’s baseline, trials short exercises (like 30-second breathing or stretches), measures impact, and ranks each by effectiveness. During ongoing use, when stress exceeds the personalized threshold, NeuroCalm suggests the top-performing technique and updates rankings dynamically via a multi-armed bandit algorithm with ε-greedy exploration. Built with React, Firebase, and Python, the platform ensures real-time data handling, secure storage, and continuous personalization. We’re proud to have implemented live stress detection, adaptive learning, and an empathetic user interface that prioritizes calm over complexity. Through this project, we learned the importance of balancing data-driven personalization with human-centered design. Next, we plan to expand NeuroCalm’s technique library, integrate wearable data for higher accuracy, and pilot the system in classrooms to support more focused, resilient learning experiences.